EBA Consults to Revise Standards on Benchmarking of Internal Models

The key update of the 2021 implementing technical standards relates to the introduction of the IFRS 9 benchmarking templates. In line with the staggered approach communicated by EBA in the IFRS 9 roadmap, in this first phase of the exercise, the IFRS 9 templates solely collect data on low-default portfolios, or LDP, with a focus on the probability of default or PD. The objective of the exercise is to collect quantitative data on the IFRS 9 Expected Credit Loss (ECL) parameters and other relevant information that, combined with a qualitative questionnaire to be filled by the institutions separately, will be used to gain a deeper understanding of the different methodologies, models, and scenarios that could lead to material inconsistencies in ECL outcomes, affecting own funds and regulatory ratios.

The initial focus on low-default portfolios is expected to allow an analysis of ECL without undue variability. It should create insights into the value of IFRS 9 parameters to which institutions have common exposures. Some additional IFRS 9 parameters will be collected for this purpose (for example, probability of default under IFRS 9 by counterparty and by economic scenario or facility). In the first stage of the exercise, these new parameters will focus on probability of default on three different aspects of the accounting framework:

• The analysis of the variability of the 12 months—probability of default parameters
• The analysis of the variability of the macroeconomic forecasts and the interaction between the lifetime probability of default curve and the macro-economic scenarios
• The analysis of variability of practices in the Significant Increase of Credit Risk (SICR) assessment 

These draft implementing technical standards have been developed in accordance with Article 78 of the Capital Requirements Directive (CRD), which requires EBA to specify the benchmarking portfolios, templates, and definitions to be used as part of the annual benchmarking exercises. These are used by competent authorities to conduct an annual assessment of the quality of internal approaches used for the calculation of own funds requirements. The EBA supervisory benchmarking already served three major objectives, the first one being the supervisory assessment of the quality of internal approaches. It also provides a powerful tool to explain and monitor RWA variability over time and the resulting implications for prudential ratios. In this role, it triggered among others the development of EBA guidelines on probability of default and loss given default estimation and the treatment of defaulted assets, which were published on the November 17, 2017. The benchmarking results also provide banks with valuable information on their risk assessment compared to other banks assessment on comparable portfolios.

These three objectives are better achieved with the integration of IFRS 9 benchmarking templates, where the most relevant sources of variability arising from the implementation of the new accounting standard and the respective consequences in terms of prudential ratios could be identified and monitored. Given the commonalities between internal ratings-based models for credit risk and IFRS 9 models, it is deemed appropriate to use the current benchmarking tool and, therefore, to build on the existing implementing technical standards on supervisory benchmarking in conducting the IFRS 9 benchmarking exercise. For this reason, changes are being suggested to Regulation 2016/2070 to integrate in the current set of templates additional templates dedicated to IFRS 9, for collecting information in terms of parameters to measure ECL and other relevant information.

Related Links

• Press Release
• Consultation Paper (PDF)
• Benchmarking Package (ZIP)
• Past Benchmarking Exercises
• IFRS 9 Roadmap, July 2019 (PDF)

Keywords: Europe, EU, Banking, Basel III, Reporting, Benchmarking Internal Models, Market Risk, Credit Risk, IFRS 9, 2021 Benchmarking Exercise, Expected Credit Loss, EBA